Air temperature retrieval from remote sensing data based on thermodynamics

A new approach to retrieving air temperature from land surface temperature is presented. The new method is based on thermodynamics. Two important parameters, namely crop water stress index and aerodynamic resistance, were used to build a quantitative relationship between the land surface temperature and the ambient air temperature. The method was applied using MODIS satellite data for a location situated in the North China Plain. Comparing the measurement values at meteorological stations with air temperature, derived by the method for certain pixels, indicates that derived values can be obtained within an accuracy of 3 C for more than 80% of data processed. Sensitivity studies also suggest that inaccuracies associated with measurement error in the model variables are also within the 3 C range.